Nov 13, 2018 · A typical measure to evaluate online learning algorithms is regret but such standard definition of regret is intractable for nonconvex models ...
The performance of online learning algorithms is commonly evaluated by regret, which is defined as the difference between the real cumulative loss and the ...
This work introduces another definition of regret, inspired by the concept of calibration and a local gradient based regret, and discusses why this ...
Nov 28, 2018 · We consider an online learning process to forecast a sequence of outcomes for nonconvex models. A typical measure to evaluate online ...
A typical measure to evaluate online learning algorithms is regret but such standard definition of regret is intractable for nonconvex models even in offline ...
We introduce a local regret for non-convex models in a dynamic environment. We present an update rule incurring a cost, according to our proposed local regret, ...
In this paper, we consider online learning with non-convex loss functions. Similar to Besbes et al. [2015] we apply non-stationary regret as the performance ...
Missing: Local | Show results with:Local
This paper studies computationally tractable notions of regret minimization and equilibria in non-convex repeated games. Efficient online learning algorithms ...
This paper introduces a local regret for non-convex models in a dynamic environment. The authors present an update rule incurring a cost that is sublinear in ...
This paper develops a methodology for regret minimization with stochastic first-order oracle feedback in online, constrained, non-smooth, non-convex ...